Download ePaper

research into the mechanisms of paraquat-induced multiple organ ...

research into the mechanisms of paraquat-induced multiple organ ... research into the mechanisms of paraquat-induced multiple organ ...

from repositorio.aberto.up.pt More from this publisher

11.12.2012 Views

� ��

�

��

�

� �

� � �

� ��

��

�

��

��

RESEARCH INTO THE MECHANISMS OF

PARAQUAT- INDUCED MULTIPLE ORGAN FAILURE

Development and application of antidotes and antidotal

pathways for the treatment of human poisonings

Ricardo Jorge Dinis Oliveira

Porto, 2007

FACULTAD DE FARMACIA

UNIVERSIDAD DE SALAMANCA

Dissertação de candidatura ao

grau de Doutor em Toxicologia apresentada

à Faculdade de Farmácia da

Universidade do Porto

Dissertation thesis for the degree

of Doctor of Philosophy in Toxicology

submitted to the Faculty of

Pharmacy of Porto University

Orientador: Professor Doutor Félix Dias Carvalho (Professor Associado com

Agregação da Faculdade de Farmácia da Universidade do Porto;

Co-orientadora: Professora Doutora Maria de Lourdes Pinho de Almeida Souteiro

Bastos (Professora Catedrática da Faculdade de Farmácia da

Universidade do Porto);

Co-orientadora: Professora Doutora Amparo Sánchez Navarro (Professora

Associada com Agregação da Faculdade de Farmácia da

Universidade de Salamanca).

iii

Aos meus Pais, José e Irene, e Irmã, Carla.

To my Parents, José and Irene, and Sister, Carla.

À Sofia.

To Sofia.

vii

viii

AUTHOR’S DECLARATION

Under the terms of the Decree-Law nº 216/92, of October 13th, is hereby declared

that the following original articles were prepared in the scope of this dissertation.

Theoretical Background

PUBLICATIONS

Articles in international peer-reviewed journals

I. Dinis-Oliveira, R. J., Remião, F., Duarte, J. A., Sanchez-Navarro, A., Bastos, M. L.,

and Carvalho, F. (2006). Paraquat exposure as an etiological factor of Parkinson’s

disease. Neurotoxicology, 27, 1110-1122.

II. Dinis-Oliveira, R. J., Remião, F., Duarte, J. A., Sanchez-Navarro, A., Bastos, M. L.,

and Carvalho, F. (2007). Paraquat poisonings: mechanisms of lung toxicity, clinical

features and treatment. (Submitted for publication).

Original Research

I. Dinis-Oliveira, R. J., Jesus-Valle, M. J., Bastos, M. L., Carvalho, F., and Sanchez-

Navarro, A. (2006). Kinetics of paraquat in the isolated rat lung. Influence of sodium

depletion. Xenobiotica 36, 724-737.

II. Dinis-Oliveira, R. J., Sarmento, A., Reis, P., Amaro, A., Remião, F., Bastos, M. L.,

and Carvalho, F. (2006). Acute paraquat poisoning: report of a survival case following

intake of a potential lethal dose. Pediatr Emerg Care 22, 537-540.

III. Dinis-Oliveira, R. J., Remião, F., Duarte, J. A., Sanchez-Navarro, A., Bastos, M. L.,

and Carvalho, F. (2006). P-glycoprotein induction: an antidotal pathway for paraquatinduced

lung toxicity. Free Radic Biol Med 41, 1213-1224.

IV. Dinis-Oliveira, R. J., Duarte, J. A., Remião, F., Sanchez-Navarro, A., Bastos, M. L.,

and Carvalho, F. (2006). Single high dose dexamethasone treatment decreases the

pathological effects and increases the survival rat of paraquat-intoxicated rats.

Toxicology 227, 73-85.

V. Dinis-Oliveira, R. J., Sousa, C., Remião, F., Duarte, J. A., Sanchez-Navarro, A.,

Bastos, M. L., and Carvalho, F. (2007). Full survival of paraquat-exposed rats after

treatment with sodium salicylate. Free Radic Biol Med 42, 1017-1028.

VI. Dinis-Oliveira, R. J., Sousa, C., Remião, F., Duarte, J. A., Sanchez-Navarro, A.,

Bastos, M. L., and Carvalho, F. (2007). Effects of sodium salicylate in the paraquatinduced

apoptotic events in rat lungs. Free Radic Biol Med 43, 48-61.

Original Research

x

Abstracts in international peer-reviewed journals

I. Dinis-Oliveira, R. J., Jesus-Valle, M. J., Bastos, M. L., Carvalho, F., and Sanchez-

Navarro, A. (2005). Influence of sodium depletion on the kinetics of paraquat in the

isolated rat lung. Toxicol Lett 158 (Suppl. 1): S213.

II. Dinis-Oliveira, R. J., Sarmento, A., Reis, P., Amaro, A., Remião, F., Bastos, M. L.,

and Carvalho, F. (2005). Acute paraquat poisoning: report of a survival case following

intake of a potential lethal dose. Toxicol Lett 15 (Suppl. 1): S241.

III. Dinis-Oliveira, R. J., Remião, F., Duarte, J. A., Sanchez-Navarro, A., Bastos, M. L.,

and Carvalho, F. (2006). A new and vital antidotal pathway for paraquat poisonings

more than 60 years later: Induction of lung P-glycoprotein. Toxicol Lett 164 (Suppl. 1):

S75.

IV. Dinis-Oliveira, R. J., Duarte, J. A., Remião, F., Sanchez-Navarro, A., Bastos, M. L.,

and Carvalho, F. (2006). Dexamethasone treatment decreases the pathological effects

and increases the survival rate of paraquat-intoxicated rats. Toxicol Lett 164 (Suppl. 1):

S237-S238.

V. Dinis-Oliveira, R. J., Remião, F., Sanchez-Navarro, A., Duarte, J. A., Bastos, M. L.,

and Carvalho, F. (2007). Recent developments in the therapy of paraquat poisoning.

Clin Toxicol (Phila) 45:376 (invited keynote presentation).

Original Research

Patents

I. Dinis-Oliveira, R. J., Remião, F., Duarte, J. A., Sanchez-Navarro, A., Bastos, M. L.,

and Carvalho, F. The use of the process of induction of de novo synthesis of Pglycoprotein

(P-gp) in the treatment of xenobiotic-induced intoxications in mammals.

Portuguese Patent Nº 103420/06 (2006) and International Patent Nº

PCT/IB2007/050144 (2007).

II. Dinis-Oliveira, R. J., Remião, F., Duarte, J. A., Sanchez-Navarro, A., Bastos, M. L.,

and Carvalho, F. (2006). The use of salicylate as an antidote of paraquat intoxications in

humans. Portuguese Patent Nº 103480/06 (2006) and International Patent

Nº PCT/IB2007/051799 (2007).

Under the terms of the referred Decree-Law, the author declares that he afforded

a major contribution to the conceptual design and technical execution of the work,

interpretation of the results and manuscript preparation of the published articles

included in this dissertation.

xii

Programa Operacional da Ciência e Inovação 2010

MINISTÉRIO DA CIÊNCIA, INOVAÇÃO E ENSINO SUPERIOR

The candidate performed the experimental work with a doctoral fellowship

(SFRH/BD/13707/2003) supported by the “Fundação para a Ciência e a Tecnologia”,

which also participate with grants to attend in international meetings and for the

graphical execution of this thesis. The Faculty of Pharmacy of the University of Porto

(Portugal) and of Salamanca (Spain) provided the facilities and logistical supports.

ACKNOWLEDGMENTS

There are lots of people I would like to thank for a huge variety of reasons. Despite

this dissertation represents an individual work, it involved more than three years of

cooperation between many people. It is a pleasure to express my truthful gratitude to all

who made this thesis possible with words of encouragement. This is perhaps the easiest

and hardest chapter that I have to write. It will be simple to name all the people that

helped to get this done, but it will be tough to thank them enough. I will nonetheless

try…

Firstly and at HEAD, I would like to thank my Supervisor, Prof. Félix Dias

Carvalho for his guidance and support during the course of this work. I could not have

imagined having a better advisor and mentor for my PhD, and without his commonsense,

knowledge, and perceptiveness I would never have finished. It is difficult to

overstate my gratitude. With his enthusiasm, his inspiration, and his great efforts to

explain things clearly and simply, he helped me to make the investigation fun for me.

Throughout my thesis period, he provided encouragement, sound advice, good teaching,

good company and friendship, and lots of good ideas. I would have been lost without

him. It has been a privilege to be able to learn from his example as a researcher and

mentor. I have been extremely lucky to have a supervisor who cared so much about my

work, and who responded to my questions and queries so promptly. Definitely, you are

an example to follow. His careful support was highly constructive for my growth and

makes him a true friend.

I would like to extend my gratitude to Prof. Maria de Lourdes Bastos, my cosupervisor,

for her help and for much type of facilities she provided me during all these

approximately three years. Her dynamic and prompt way of being and her work

capacity are remarkable characteristics to pursue. With all respect, her “youth” of spirit

is an example of life.

I am very grateful to Professor Amparo Sánchez Navarro, my co-supervisor, for

giving me the opportunity to work at Department of Pharmacy and Pharmaceutical

xiii

Technology, Faculty of Pharmacy of University of Salamanca, for her welcoming

during my stays in Salamanca, valuable guidance, numerous critical appraisals and

scientific discussions, not always easy because of the distance. The answers to my email

questions and the comments on my manuscripts were always sent back sooner than

I expected.

I'm grateful to my “supervisor” Professor José Alberto Duarte from CIAFEL of

the Faculty of Sport of the University of Porto. Our discussions throughout the nights

would certainly be important to my constant academic growth. His arduous, rigorous,

meticulous work and valuable help in many steps was of incontestable value. I am

deeply indebted to Professor José Alberto Duarte for his involvement, unlimited

contribution to this study, for the many insightful conversations during the development

of new ideas, and for helpful comments on the text. I want to express that he was an

outstanding advisor and an excellent professor. I could always count with him, and I

will always be grateful for his expertise and support. When I met him for the first time I

was far from realise that it would certainly be very difficult to conclude this thesis

without his friendship. Thanks for all Professor.

I’m grateful to Prof. Fernando Remião for his assistance in helping to supervise

me, providing resources and subjects, offering direction and penetrating criticism, for

many insightful conversations during the development of new ideas in this thesis, for

the fun environment he provided, and for making the COHiTEC project possible. I’m

also thankful to have benefited from his previous works on paraquat quantification and

for all the remarkable worries he demonstrated in supplying good laboratory conditions.

Professor, I am grateful in every possible way and hope to keep up our collaboration in

the future with our entrepreneur life style.

I acknowledge “Fundação para a Ciência e a Tecnologia” for my doctoral

fellowship (SFRH / BD / 13707 / 2003) and for the financial support of this dissertation.

Thanks for the contribution.

I acknowledge “REQUIMTE”, associated laboratory, for the financial support of

the laboratory work.

xiv

To Engineer Elisa, the Mum of the department, besides the important life lessons

she gave me, I’m thankful for her friendship and companionship forever.

To all my toxicology department friends, I would like to thank their laboratorial

and fieldwork assistance as well as their companionship and the permanent support and

motivation. In particular, my appreciation to Carla, Cecília, Estela, Helena Carmo,

Helena Pontes, João, Renata, Teresa e Vera, for their friendship and fun environment in

which I learnt, grew and for the helpful support along these years. A special encourage

word to Carla for her irreplaceable and exceptional support during all laboratorial

assays. She really has an appreciable work capacity in the laboratory. To Görkem

Mergen for all the entertaining moments I send you a big hug.

To Mrs. Julia, Graziela and Conceição for their prompt support whenever

necessary in solving technical problems, for the careful washing and handling with the

laboratory material and for their efficient resolution of logistic problems. We all miss

Mrs. Graziela very much in the Faculty. Wherever you are now, we hope you are in

peace... Thanks for all the help.

To Mrs. Casemira for her prompt, gentleness and sympathy in provide missing

laboratory material to my work persecution.

I would like to thank to my high school chemistry teacher (Professor Augusta

Silveira) for the kind assistance, advices and help in conferring some justice to the

marks and consequently to make everything possible. A special thank for that Professor.

A special thank to David Costa for all the help and knowledge he shared without

asking anything to exchange. You are a good guy.

I’m gratified to Professor Eduarda Fernandes for her collaboration in the

persecution of the work, mainly helping me leading with the Nuclear magnetic

resonance (NMR) spectrums.

I’m grateful to Professor Franklim Marques for had provided the facilities in the

beginning of the research.

A special word of acknowledge to Professor Natércia, to Professor Georgina and

to Margarida for their collaboration leading with the TUNEL experiments.

I’m thankful to Ana Margarida for her amiable support regarding to animals’ care

and treating protocols.

To Rita Ferreira from CIAFEL who participated and well contributed in several

biochemical quantifications.

To Mrs. Celeste Resende for her technical assistance regarding the samples

processing for Light and Electron Microscopy. Naturally, I apologise for the time I still

you.

Doctor Cândida, the librarian, I am grateful to her assistance in providing the

bibliography and for that she deserve special mention.

To all my friends from Department of Pharmacy and Pharmaceutical Technology

of the Faculty of Pharmacy of the University of Salamanca for his friendship in all

moments we shared together. A special thank for the beginning of my scientific

edification.

I thankfully acknowledge the promptness help and friendly contribution of my

friends Ricardo Silvestre, Joana Maciel and Joana Tavares from the Department of

Biochemistry.

To Sofia, my girl-friend, for her sincerely friendship, constant care in good and

bad moments, emotional support, camaraderie, entertainment, and caring she provided. I

would like to wish you the same you desire for me. She was my own "soul out of my

soul," who kept my spirits up when the muses failed me.

To my lovely Mum, Maria Irene, and Dad, José Carneiro, and to my sister, Carla

Isabel, who experienced all of the ups and downs of my research and for all the

patience, to them I dedicate this thesis. Despite far away most of the times I know that

we are always close. You are unique and I’m very proud for being your sun and brother.

xvi

To my godfather, José Pinto, and to Paula and Tó for all the advices in anxiety and

apprehension moments, my special acknowledge.

To finalize I want to give a word for the loving memory of my godmother and

grandparents, Joaquim Dinis and Severino da Costa. I miss you so much…

Is that everyone?

Ricardo Dinis, Porto, 2007

xvii

xviii

ABSTRACT

Paraquat (PQ) is a popular herbicide and it is probably one of the most studied

pesticides. This interest is related not only to its relevance as a human poison, but also

to its specific toxicological mechanisms, namely in the area of oxidative stress, making

it an excellent research tool. PQ has a proven safety record when used properly for its

intended purpose. Its safety in use can, at least in part, be explained by its lack of

absorption either by inhalation or through the intact skin. This is because the spray

droplets generated by agricultural equipment are too large in diameter (>5 μm) to be

inhaled and because the skin provides an effective, impermeable barrier to the

absorption of PQ. However, over the past 44 years, there have been numerous fatalities

following accidental or deliberate ingestion of this weed-killer. Poisonings have

received considerable attention in the medical, scientific, and popular press. The

intoxication cases have often been dramatic because of the protracted and inexorable

course of the illness and the absence of any effective antidote. Fatality rates have been

over 50%. In addition, very little pre-clinical research has been successful in applying

antidotes that shown to work in animal studies into clinical practice. PQ accumulates

mainly in the lung (pulmonary concentrations can be six to ten times higher than those

in the plasma), where it is retained even when blood levels start to decrease. The

pulmonary effects can be readily explained by the participation of the polyamine

transport system abundantly expressed in the membrane of alveolar cells type I, II and

Clara cells. Further downstream at the toxicodynamic level, the molecular mechanism

of PQ toxicity is based on redox cycling and intracellular oxidative stress generation.

During the last years, our research group has been a reference in the field of PQ

toxicity to hospitals in the centre and north of Portugal. According, this dissertation was

primarily aimed to describe a successful clinical case, regarding the intoxication of a 15year-old

girl by a presumed lethal dose of PQ. Besides the measures for decreasing PQ

absorption and increasing its elimination, other protective procedures were applied

aiming to reduce the production of reactive oxygen species (ROS), scavenge and repair

ROS-induced lesions, and to reduce inflammation. The status-of-the-art concerning the

biochemical and toxicological aspects of PQ poisoning and the pharmacological basis of

the respective treatment protocol was presented. It was conducted an intensive and

xix

aggressive treatment, based on the high urinary and plasmatic PQ concentrations, and

accordingly to the positive outcome, the therapeutic protocol followed could be a

promising treatment of PQ human intoxications. Besides, were also good prognostic

factors, young age, lesser degrees of leukocytosis and acidosis, and the absence of renal,

hepatic, and pancreatic failures on admission after acute PQ poisoning.

xx

In this thesis, the usefulness of the isolated rat lung was firstly explored. Such

model was applied to characterize the toxicokinetic behaviour of PQ in this tissue after

bolus injection under standard experimental conditions as well as to evaluate the

influence of iso-osmotic replacement of sodium by lithium in the perfusion medium.

The obtained results showed that the isolated rat lung model is a useful technique for

PQ toxicokinetic studies. It was also observed that sodium-depletion in the perfusion

medium leads to a decreased uptake of PQ in the isolated rat lung although it seems that

this condition does not contribute to improve the elimination of PQ once the herbicide

reaches the extravascular structures of the tissue. In spite of, techniques of tissue

isolation and perfusion offer an excellent alternative to characterize the kinetic profile

for a tissue in a single animal and avoids the inter-individual variability in each single

curve, leading as well to a corresponding reduction in curve replicates and hence a

substantial reduction in the number of animals used (5-8 versus 50-80/tissue), soon we

notice that PQ toxicity is a myriad of factors, together contributing to a death outcome

and it would be better studied by in vivo approaches.

According to this desideratum, secondly it is described a procedure, through the

induction of de novo synthesis of P-glycoprotein by the administration of a single high

dose of dexamethasone (DEX) to Wistar rats, that leads to a remarkable decrease of PQ

accumulation in the lung, together with an increase of its faecal excretion and a

subsequent decrease of several biochemical and histopathological biomarkers of

toxicity. The obtained results shown that DEX also ameliorated the biochemical and

histological liver alterations induced by PQ in Wistar rats. On the other hand, these

improvements were not observed in kidney and spleen of DEX treated rats. The sum of

these effects was clearly positive, since it was observed an increased survival rate,

which indicates that high dosage DEX treatment constitutes an important and valuable

therapeutic tool to be used against PQ-induced toxicity.

Finally the role of the apoptosis, oxidative stress, platelet aggregation, nuclear

factor (NF)-κB activation and fibrosis in PQ-induced lung toxicity, as well as the

remarkable healing effects obtained by the administration of sodium salicylate (NaSAL,

200 mg/Kg i.p.), were assessed. The obtained results exceeded our best expectations

since not only the toxicity was reverted but, most significantly, full survival of the PQ-

intoxicated rats treated with NaSAL was observed. It may be postulated that NaSAL is

the first real PQ antidote described with such degree of success.

Of note, the administrations of DEX and NaSAL were given two hours after

intoxication of rats with PQ, a lag time that confers realism to be applied in humans,

since this chronological time corresponds to longer biological time for humans and

therefore this represent the probable time that passes between the herbicide ingestion

and the begin of the medical cares.

In conclusion, the results of this dissertation suggest that high doses of DEX

and/or NaSAL are therapeutic approaches with potential to be applied in humans,

though, further pre-clinical studies are needed particularly those aimed to explain in

more detail the mode of action of these interesting drugs in the protection against PQinduced

lung damage.

xxi

xxii

RESUMO

O paraquato (PQ) é provavelmente um dos pesticidas mais estudados. Este

interesse não está apenas relacionado com a sua relevância como um xenobiótico

envolvido em intoxicações humanas, mas também devido aos seus mecanismos

toxicológicos, nomeadamente na área do stress oxidativo onde representa uma

ferramenta de extrema importância. O PQ tem uma comprovada segurança quando

usado devidamente, podendo esta ser em parte explicada pela ausência de absorção por

via inalatória e através da pele integra. Isto porque, o tamanho das gotículas produzidas

pelos equipamentos agrícolas, apresentam um diâmetro demasiadamente grande (>5

μm) para serem inaladas e também porque a pele constitui uma barreira impermeável à

absorção do PQ. No entanto, nos últimos 44 anos, o PQ tem sido a causa de diversas

mortes sobretudo por ingestão, acidental ou voluntária, mas também por exposição

dérmica. As intoxicações rapidamente receberam considerável atenção da comunidade

médica, científica e dos meios de comunicação social. Apesar das razões para o uso

deste herbicida como agente de suicídio sejam difíceis de determinar, o principal

responsável parece ser estes últimos que informam e documentam os suicídios

resultantes da toxicidade aguda do PQ.

Os casos de intoxicação têm sido dramáticos, muito porque o curso da doença é

rápido e pela total ausência de um antídoto ou tratamento eficaz, dependendo a

sobrevivência dos intoxicados da quantidade ingerida e do tempo que decorre até o

início das intervenções médicas para eliminar o PQ que ainda não foi absorvido e/ou

captado pelas células. As taxas de letalidade cifram-se em valores acima dos 50%. Além

do mais, muito pouco do resultante da investigação pré-clínica tem sido aplicado com

sucesso na apática clínica. O PQ acumula-se maioritariamente no pulmão, onde as

concentrações podem atingir seis ou mesmo dez vezes as concentrações plasmáticas,

ficando aí mesmo quando os níveis plasmáticos começam a diminuir. Os efeitos

pulmonares podem ser facilmente explicados pela participação do transportador das

poliaminas endógenas abundantemente expressado na membrana os pneumócitos Tipo

I, II, e nas células Clara. Ao nível toxicodinâmico, o mecanismo de toxicidade do PQ é

baseado no ciclo redox e na geração de espécies reactivas do oxigénio (ROS).

xxiii

Durante os últimos anos, o nosso grupo de investigação tem sido uma referência

para hospitais no centro e norte de Portugal no que se refere à toxicidade do PQ. Em

conformidade, o primeiro objectivo desta dissertação foi descrever um caso clínico de

sucesso, de uma jovem de 15 anos de idade que voluntariamente ingeriu uma dose letal

de PQ. Para além de medidas destinadas a diminuir a absorção do PQ ou aumentar a sua

eliminação, outras medidas protectoras foram também seguidas de modo a diminuir a

produção das ROS, captar as ROS, reparação das lesões produzidas pelas ROS e

diminuir a inflamação. O estado da arte referente aos aspectos bioquímicos e

toxicológicos do PQ e a base farmacológica do protocolo terapêutico seguido foi

apresentado. Conduziu-se um tratamento intensivo e agressivo baseado nos altos níveis

urinários e plasmáticos e tendo em conta o resultado positivo, foi possível concluir que

o protocolo seguido poderá ser prometedor no tratamento das intoxicações humanas

pelo PQ. Foram também factores de prognósticos favoráveis, a juventude, menores

graus de leucocitose, ausência de falha renal, hepática e pancreática à admissão após

intoxicação aguda.

De forma a reduzir a morbilidade e mortalidade associada às intoxicações pelo

PQ, primeiramente nesta dissertação, explorou-se a utilidade do modelo de pulmão

isolado de rato com o objectivo de caracterizar o comportamento toxicocinético do PQ

neste tecido, após injecção por bólus sob condições padrão assim como para avaliar a

influência da substituição iso-osmótica do sódio pelo lítio do meio de perfusão. Os

resultados obtidos comprovaram a aplicabilidade do modelo de pulmão isolado no

estudo da toxicocinética do PQ. Observou-se também que a depleção de sódio do meio

de perfusão conduziu a diminuição da captação pulmonar de PQ neste modelo de

estudo, apesar desta condição não levar a um aumento da eliminação do PQ deste

tecido, uma vez alcançado as estruturas extracelulares. Apesar das técnicas de

isolamento e perfusão de órgãos oferecerem uma excelente alternativa para a

caracterização do perfil toxicocinético em um tecido de um animal, evitando as

diferenças de variabilidade interindividuais em cada curva, conduzindo também a uma

redução dos replicados da curva e como tal uma redução do numero de animais usados

(5-8 versus 50-80/tecido), rapidamente se constatou que a toxicidade do PQ resulta de

uma miríade de factores, juntos contribuindo para morte, os quais seriam melhor

estudados usando uma abordagem in vivo.

De acordo com este desiderato, secundariamente nesta dissertação é descrito um

procedimento, através da indução da síntese de novo da glicoproteína-P (P-gp) por

xxiv

administração de uma dose única mas elevada de dexametasona (DEX) a ratos Wistar, o

qual conduziu a uma marcada diminuição da acumulação pulmonar do PQ para menos

de 40% em apenas 24 horas, associado a um aumento da sua excreção fecal e

consequente melhoramento de vários parâmetros bioquímicos e histopatológicos de

toxicidade. Os resultados obtidos demonstraram que o tratamento com DEX também

produziu uma melhoria nas alterações dos mesmos parâmetros no fígado de animais

intoxicados pelo PQ. Apesar de tais resultados positivos não terem sido registados no

baço e no rim após tratamento com DEX, o somatório destes efeitos foi claramente

benéfico, uma vez que foi observado um aumento da percentagem de sobrevivência

para 50% ao final de 10 dias em oposição aos 10% reportados em estudos anteriores

realizados por outros grupos de investigação e utilizando múltiplas abordagens

terapêuticas. Pode-se desta forma concluir que altas doses de DEX constituem uma

importante e valiosa ferramenta terapêutica a ser usada nas intoxicações pelo PQ.

Apesar da extrema importância deste estudo, a persistente lacuna relacionada com

a inexistência de um antídoto que garanta 100% de sobrevivência das intoxicações pelo

PQ motivou finalmente o último estudo documentado nesta dissertação, no qual se

demonstrou que o salicilato de sódio (NaSAL, 200 mg/Kg i.p.) tem um elevado

potencial para constituir um verdadeiro antídoto das intoxicações pelo PQ. Neste estudo

foi avaliado o papel da apoptose, do stress oxidativo, da activação plaquetária, do factor

de transcrição pro-inflamatório, Nuclear Factor (NF)-κB e da fibrose, assim como os

remarcáveis efeitos protectores do NaSAL na toxicidade pulmonar induzida pelo PQ.

Verificou-se que a alteração destes factores como consequência da exposição ao PQ,

pode ser significativamente inibida pelo NaSAL com a consequente recuperação dos

animais intoxicados. Na verdade, este estudo revelou-se muito mais importante do que à

partida era esperado, pois pela primeira vez, foi demonstrado que com este fármaco e

com apenas uma dose foi possível reverter toda a toxicidade dos animais intoxicados

com o PQ e garantir uma percentagem de sobrevivência de 100%.

De salientar que o efeito antidotal da DEX e do NaSAL resultou da administração

destes fármacos 2 horas após a intoxicação dos animais com o PQ. Este intervalo

confere um maior realismo na aplicação às intoxicações humanas, pois reflecte, em

grande parte dos casos, o tempo que medeia entre a ingestão deste herbicida e o início

do possível tratamento hospitalar do paciente.

Em conclusão, os resultados desta dissertação sugerem que altas doses de DEX

e/ou de NaSAL constituem importantes abordagens terapêuticas com potencial

xxv

aplicativo em humanos, apesar de mais estudos pré-clínicos serem necessários,

nomeadamente aqueles destinados a clarificar modo de acção do NaSAL na prevenção

da lesão pulmonar originada pelo PQ e também aqueles com o objectivo de encontrar

novos, específicos e mais potentes indutores da síntese de novo da P-gp, isentos dos

efeitos adversos que são bem conhecidos para os glucocorticóides.

xxvi

Ab, antibody;

ACE, angiotensin-converting enzyme;

AP-1, activator protein-1;

ARDS, acute respiratory distress syndrome;

ATP, adenosine triphosphate;

BALF, bronchoalveolar lavage fluid;

BHs, bipyridylium herbicides;

BW, body weight;

CHP, charcoal hemoperfusion;

CLCr, creatinine clearance;

CLPQ, paraquat clearance;

Cmax, maximum plasma concentration;

CNS, central nervous system;

CP, cyclophosphamide;

CP51, hydroxypyridin-4-one;

Cyt c, Cytochrome c;

DEX, dexamethasone;

DFO, desferoxamine;

ABBREVIATIONS LIST

DLCO, lung carbon monoxide diffusing capacity;

DNA, deoxyribonucleic acid;

ERG, electroretinogram;

Fe 2+ , ferrous ion;

Fe 3+ , ferric ion;

FiO2, fraction of inspired oxygen;

FR, Fenton reaction;

FRD, ferrodoxin;

G6PD, glucose-6-phosphate dehydrogenase;

GFR, glomerular filtration rate;

GGO, ground-glass opacification;

GIT, gastrointestinal tract;

xxvii

GPx, glutathione peroxidase;

Gred, glutathione reductase;

GSH, reduced glutathione;

GSSG, oxidized glutathione;

H2O2, hydrogen peroxide;

HMP, hexose monophosphate pathway;

HO . , hydroxyl radical;

HRCT, high-resolution computed tomography;

HWR, Haber-Weiss reaction;

i.p., intraperitoneal;

i.v., intravenous;

ICI, Imperial Chemical Industries (now Syngenta);

Km, Michaelis-Menten constant;

LPO, lipid peroxidation;

MDA, malondialdehyde;

MINA, 4-methylisonicotinic acid;

MGBG, methylglyoxal bis-(guanylhydrazone);

MP, methylprednisolone;

Na + , sodium;

NAC, N-acetylcysteine;

NADP + , oxidized nicotinamide adenine dinucleotide phosphate;

NADPH, reduced nicotinamide adenine dinucleotide phosphate;

NaSAL, sodium salicylate;

NF-κB, nuclear factor kappa-B;

NMN, N-methylnicotinamide;

NO, nitric oxide;

NOS, nitric oxide synthase;

O2, oxygen;

O2 .- , superoxide radical;

PaCO2, partial pressure of carbon dioxide in arterial blood;

PAH, p-aminohippurate;

PaO2, partial pressure of oxygen in arterial blood;

PAO2, partial pressure of oxygen in the alveolus;

PEEP, positive end-expiratory pressure;

xxviii

PFTs, pulmonary function tests;

P-gp, P-glycoprotein;

PQ or PQ 2+ , paraquat;

PQ •+ , paraquat monocation free radical;

PUS, polyamine uptake system;

RNA, ribonucleic acid;

ROS, reactive oxygen species;

s.c., subcutaneous;

SH, thiol;

SOD, superoxide dismutase;

t1/2, half-life;

Tmax, time to maximum plasma concentration;

TPC, TUNEL-positive cells;

TUNEL, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate

nick end-labeling;

VER, verapamil;

Vmax, maximal rate;

WBC, white blood cell;

XD, xanthine dehydrogenase;

XO, xanthine oxidase.

xxix

xxx

INDEX OF FIGURES

Fig. 1 – Chemical structures of paraquat (PQ) and diquat (DQ)...................................... 3

Fig. 2 – Equilibrium dynamics for paraquat between the soil and soil solution. Adapted

from Roberts et al. (Roberts et al., 2002)......................................................................... 6

Fig. 3 – Herbicidal mechanism of paraquat. In photosystem I (PS I), plastocyanin (PC)

transfers its electron (e - ) through a series of steps (P700, A0, A1, FX, FA) to ferrodoxin

(FRD) and finally to NADP + . PQ ion (PQ 2+ ) binds near the FRD binding site in PS I and

accepts an e - , becoming paraquat monocation free radical (PQ •+ ), which initiates a series

of reactions leading to cell membrane disruption and plant death. The formation of such

free radicals stops electron transport to NADP + and effectively inhibits normal

functioning of PS I. A, Ferrodoxin-NADP + reductase. .................................................... 8

Fig. 4 – Photochemical and microbial degradation of PQ. Adapted from Slade (Slade,

1965)............................................................................................................................... 10

Fig. 5 - Synthesis of paraquat......................................................................................... 14

Fig. 6 - Paraquat dealkylation in alkaline solutions. ...................................................... 15

Fig. 7 – Intermediate resonance structures of paraquat and full reduction. ................... 16

Fig. 8 - Chemical structure of paraquat (A) and putrescine (B), showing geometric

standards of the distance between N atoms (optimal distance to fit polyamine uptake

system is unknown). The chemical structure of cadaverin (C), spermidine (D) and

spermine (E) is also presented........................................................................................ 25

Fig. 9 - Autoradiographs of rat lung tissue incubated with [ 3 H]putrescine. Resin sections

1 µm thick were stained with toluidine blue and examined by light microscopy.

Labeling occurs in alveolar walls and in alveolar type II pneumocytes (A and B, arrows).

There is no labeling in macrophages (B) or in walls of vessels, but Clara cells

xxxi

(arrowheads) in bronchiolar epithelium (C) show intense labeling. Original

magnifications: ×600 in A; ×1,500 in B and C. Adapted from Nemery et al. (Nemery et

al., 1987)......................................................................................................................... 29

Fig. 10 - Autoradiographs of human lung tissue incubated with 2.5 µM [ 3 H]putrescine.

Unstained resin sections 1 µm thick were examined by electron spectroscopic imaging.

a-d: 4 different alveolar spaces lined with type II and type I pneumocytes. Silver grains

are evident over type II pneumocytes (long arrows) and lining of alveoli (short arrows)

but not over erythrocytes (*) or paranuclear regions of endothelium. Cellular and

noncellular components of alveolar interstitium were largely devoid of silver grains.

Silver grains were uniformly distributed over both nucleus and cytoplasm of type II

cells. Bars, 1 µm. Adapted from Hoet et al. (Hoet et al., 1993)..................................... 30

Fig. 11 - Schematic representation of the mechanism of paraquat toxicity. A. Cellular

diaphorases, SOD, Superoxide dismutase; CAT, Catalase; GPx, Glutathione Peroxidase;

Gred, Glutathione Reductase; PQ 2+ , Paraquat; PQ •+ , Paraquat monocation free radical;

HMP, Hexose monophosphate pathway, FR; Fenton reaction; HWR, Haber-Weiss

Reaction, PUS; polyamine uptake system...................................................................... 35

Fig. 12 – Pretreatment procedures for paraquat in urine and plasma before the second

derivative spectrophotometric analysis (A). Zero-order and second-derivative spectrum.

The qualitative analysis is made by observing the presence of inflection points at about

396 and 403 nm. The quantification is made with amplitudes measurable between 396

and 403 nm (B)............................................................................................................... 63

Fig. 13 - Nomogram showing relation between plasma paraquat concentrations

(μg/mL), time after ingestion, and probability of survival. Adapted from Hart et al.

(Hart et al., 1984). .......................................................................................................... 65

Fig. 14 – Representative qualitative urinary test for paraquat. Correlation between the

PQ concentration (μg/mL) and the intensity of the blue colour change......................... 68

Fig. 15 – Relationship between urine paraquat concentrations and survival. Adapted

from Scherrmann et al. (Scherrmann et al., 1987). ........................................................ 69

xxxii

Fig. 16 - Flowchart guide usually followed for the management of paraquat poisoning.

PQ, paraquat; i.v., intravenous; PaO2, partial pressure of oxygen in arterial blood; O2,

oxygen; NO, nitric oxide; CHP, charcoal hemoperfusion; CP, cyclophosphamide; MP,

methylprednisolone; DFO, desferoxamine; NAC, N-acetylcysteine; DEX,

dexamethasone; WBC, white-blood-cells; 1 If systemic toxicity is suspected, test urine

for PQ. There is little data for time to peak plasma levels by skin absorption, but if the

urine is negative for 24 hours, systemic toxicity can probably be disregarded. If the

urine test is positive or if there is any doubt about potential systemic toxicity, assess

blood concentrations and treat for systemic toxicity as described for ingestion; 2 Risk of

inducing bleeding, perforation or scarring due to additional trauma to fragilized tissues.

Gastric lavage without administration of an adsorbent has not shown any clinical

benefit; 3 Or in 250 mL of cathartics (it will increase gut motility to improve excretion of

the charcoal-PQ complex) via nasogastric tube; 4 Maximum dose is 50 g; 5 Repeat doses

of cathartics may result in fluid and electrolyte imbalances, particularly in children, and

are therefore not recommended; 6 Particularly important as a mean to correct

dehydration, accelerating excretion, reducing tubular concentrations and correcting any

metabolic acidosis. However, fluid balance must be monitored to avoid fluid overload if

renal failure develops. In this case hemodialysis or hemofiltration may be required;

7

Plasma should be analyzed rather than serum, because serum PQ concentrations are

approximately 3 fold lower than those in plasma obtained from the same blood sample.

If only serum is available results should be interpreted with caution in relation to

survival curves. Plasma should be stored in plastic and not in glass tubes because PQ 2+

adsorb onto glass surfaces. ............................................................................................. 74

Fig. 17 – Proposed protective mechanisms of sodium salicylate against pulmonary

paraquat toxicity. .......................................................................................................... 215

xxxiii

xxxiv

INDEX OF TABLES

Table 1 – Some paraquat trade names. ............................................................................ 4

Table 2 - Countries in which paraquat is registered as of June 2005. Adapted from

paraquat information center (www.paraquat.com)........................................................... 5

Table 3 - Physical and chemical properties of PQ ion. 1 1 g of paraquat dichloride =

0.724 g of paraquat ion................................................................................................... 12

Table 4 - Kinetic constants for the accumulation of paraquat into lung tissue slices from

various species Adapted from Rose et al. (Rose et al., 1974)........................................ 23

Table 5 – Phases of paraquat toxicity and associated clinical effects. GIT,

gastrointestinal tract; CNS, central nervous system; 1 doses as low as 4 mg/Kg can cause

death (Driesbach, 1983).................................................................................................. 41

Table 6 – Paraquat LD50 in various species. NS, not stated; M, male; F, female; a dose

quoted as paraquat ion; b as dimethylsulphate................................................................. 47

Table 7 – Clinical features of paraquat poisonings........................................................ 49

xxxv

xxxvi

PART I

OUTLINE OF THE DISSERTATION

The present thesis is structured in four main parts:

1. GENERAL INTRODUCTION

In Part I, the present section, a general overview on the research assumptions,

objectives and structure of the thesis is presented. Considering the huge number of

publications concerning to PQ (more than 4000 referred in “PubMed” database since its

introduction into the market as an herbicide in 1962), the general introduction section

had a focused approach covering aspects of the clinical features of poisoning,

mechanisms of toxicity, putative treatments and their relevance to the treatment of

human poisonings. The introduction is restricted principally to those articles in high

standard scientific publications. Considerable space is dedicated to techniques for

prognosis prediction, since these could allow development of rigorous clinical protocols

that may produce comparable data for the evaluation of proposed therapies. It is

expected that this thesis may also serve as a guide for clinicians and scientists who work

and study this particularly toxic chemical.

The recent evidences linking PQ exposure with Parkinson’s disease development

are also reviewed in this section. Although the original research prepared in the scope of

this dissertation was not directed to the nervous system, the obtained findings may

provide new perspectives and foster further investigation on the involvement of PQ in

Parkinson’s disease.

2. GENERAL AND SPECIFIC OBJECTIVES OF THE DISSERTATION

The general and specific objectives of the dissertation are provided.

xxxvii

PART II – ORIGINAL RESEARCH

The Part II is divided in six chapters, corresponding to the original articles, and

describes the experimental work in order to answer the questions that derived from the

general and specific objectives of the thesis.

PART III

This section it is divided in three major points:

1. INTEGRATED OVERVIEW OF THE PERFORMED STUDIES – the

studies undertaken are integrated in a harmonized form;

2. CONCLUSIONS - the conclusions that can be taken from this dissertation are

summarized;

3. DIRECTIONS FOR FUTURE RESEARCH – future studies are projected.

PART IV

The references used in the PART I and III are listed.

xxxviii

TABLE OF CONTENTS

AUTHOR’S DECLARATION ....................................................................................VIII

PUBLICATIONS ........................................................................................................... IX

Articles in international peer-reviewed journals ............................................................ IX

Abstracts in international peer-reviewed journals............................................................X

Patents............................................................................................................................. XI

ACKNOWLEDGMENTS............................................................................................XIII

ABSTRACT ..............................................................................................................XVIII

RESUMO ..................................................................................................................XXIII

ABBREVIATIONS LIST ....................................................................................... XXVII

INDEX OF FIGURES.............................................................................................. XXXI

INDEX OF TABLES ..............................................................................................XXXV

OUTLINE OF THE DISSERTATION ................................................................ XXXVII

TABLE OF CONTENTS .......................................................................................XXXIX

PART I

(THEORETICAL BACKGROUND)

1. GENERAL INTRODUCTION ......................................................................... 1

1. HISTORY, USE AND USEFULNESS OF PARAQUAT........................................... 3

1.1 Mode of action as herbicide ................................................................................ 7

1.2 Biodegradation Pathways.................................................................................... 8

1.2.1 Photochemical Degradation.................................................................. 9

1.2.1.1 On plant surfaces.............................................................................. 9

1.2.1.2 On soil and other mineral surfaces ................................................ 10

1.2.2 Microbial degradation ........................................................................ 11

2. CHEMISTRY OF PARAQUAT ................................................................................ 11

2.1 Physical and chemical properties ...................................................................... 12

2.2 Synthesis ........................................................................................................... 13

2.3 Electrochemistry of viologens and paraquat reduction..................................... 15

3. TOXICOKINETICS OF PARAQUAT...................................................................... 17

3.1 Absorption......................................................................................................... 17

3.2 Distribution ....................................................................................................... 19

3.2.1 Preferential accumulation in the lung................................................. 22

3.2.2 Lung accumulation through the polyamine uptake system ................ 22

3.2.3 Structural requirements for the pulmonary polyamine uptake system26

3.2.4 Characterization of the pulmonary polyamine uptake system............ 27

3.2.5 Cellular localization of the polyamine uptake system in the lung...... 28

xxxix

3.3 Metabolism........................................................................................................ 31

3.4 Elimination........................................................................................................ 32

4. BIOCHEMICAL MECHANISMS OF PARAQUAT TOXICITY............................ 34

4.1 Mechanism of toxicity....................................................................................... 34

4.2 Biochemical consequences of the redox cycling process ................................. 36

4.2.1 Oxidation of NADPH ......................................................................... 36

4.2.2 Oxidation of cellular thiol (SH) groups.............................................. 37

4.2.3 Oxidative damage to lipids, proteins and DNA.................................. 38

5. LUNG PATHOPHYSIOLOGY ................................................................................. 40

5.1 Destructive phase .............................................................................................. 42

5.2 Proliferative phase............................................................................................. 44

6. OBSERVATIONS IN ANIMALS AND HUMANS ................................................. 46

6.1 Clinical symptoms and manifestations of paraquat intoxication ...................... 49

6.1.1 Poisoning by the oral route................................................................. 50

6.1.1.1 Severe toxicity................................................................................. 50

6.1.1.2 Moderate toxicity – the typical intoxication................................... 51

6.1.1.2.1 First phase................................................................................. 51

6.1.1.2.2 Second phase ............................................................................ 52

6.1.1.2.3 Third phase ............................................................................... 53

6.1.1.3 Mild Toxicity................................................................................... 55

6.1.2 Exposure by dermal route................................................................... 55

6.1.3 Ocular irritation .................................................................................. 57

6.1.4 Exposure by inhalation ....................................................................... 58

6.1.5 Muscle toxicity ................................................................................... 58

6.2 Intoxications during pregnancy......................................................................... 58

6.3 Incidents of pet animals poisoning.................................................................... 59

7. PREDICTING HUMAN OUTCOME IN PARAQUAT POISONING..................... 59

7.1 Paraquat quantification in biological samples................................................... 60

7.1.1 Qualitative and semi-quantitative test ................................................ 61

7.1.2 Quantitative test: spectrophotometry.................................................. 61

7.1.2.1 Reagents and their preparation...................................................... 62

7.4.2.2 Analytical conditions...................................................................... 62

7.1.2.3 Procedures...................................................................................... 62

7.2 Predicting the outcome from plasma paraquat concentrations.......................... 64

7.3 Predicting outcome from urine paraquat concentrations................................... 68

7.4 Additional laboratory tests ................................................................................ 69

8. TREATMENT............................................................................................................ 72

8.1 Preventing paraquat absorption......................................................................... 72

8.2 Increasing paraquat elimination ........................................................................ 76

8.2.1 Extracorporeal elimination ................................................................. 76

8.2.2 Forced diuresis and peritoneal dialysis............................................... 78

8.3 Supportive therapies.......................................................................................... 79

8.4 Measures to prevent lung damage..................................................................... 80

8.5 New perspectives .............................................................................................. 94

8.5.1 Mechanical ventilation with additional inhalation of NO .................. 95

8.5.2 Propofol .............................................................................................. 96

9. SEQUELAE IN SURVIVORS................................................................................... 96

10. LUNG APPEARANCE AT AUTOPSY .................................................................. 96

11. REVIEW ARTICLE - PARAQUAT EXPOSURE AS AN ETIOLOGICAL

FACTOR OF PARKINSON'S DISEASE...................................................................... 99

2. GENERAL AND SPECIFIC OBJECTIVES OF THE DISSERTATION ...... 115

PART II

(ORIGINAL RESEARCH)

CHAPTER I.................................................................................................... 123

CHAPTER II................................................................................................... 141

CHAPTER III.................................................................................................. 147

CHAPTER IV.................................................................................................. 161

CHAPTER V................................................................................................... 177

CHAPTER VI.................................................................................................. 191

PART III

1. INTEGRATED OVERVIEW OF THE PERFORMED STUDIES ................. 207

2. CONCLUSIONS......................................................................................... 217

3. DIRECTIONS FOR FUTURE RESEARCH ................................................ 221

PART IV

1. REFERENCES........................................................................................... 225

xli

xlii

1. GENERAL INTRODUCTION

PART I

1. GENERAL INTRODUCTION

Part I - General Introduction__________________________________________________

__________________________________________________Part I - General Introduction

1. HISTORY, USE AND USEFULNESS OF PARAQUAT

Paraquat (1,1′-dimethyl-4,4′-bipyridylium dichloride; PQ) [CAS #1910-42-5] is an

herbicide belonging to the chemical family of bipyridylium (also called bipyridyl)

quaternary ammonium herbicides. PQ and diquat (1,1′-ethylene-2,2′-dipyridylium

dibromide; DQ) [CAS #85-00-7] are the most commonly used herbicides of this group

(Fig. 1). They have similar chemical and physical properties and have a similar mode of

action on plants (Calderbank, 1968). Of the two bipyridylium herbicides (BHs) in use,

PQ is by far the most clinically significant in terms of number of intoxication cases, and

it will be the main subject of this introduction.

H 3

C

+ +

N

N CH 3

1.02 nm

PQ (1,1´-dimethyl-4,4´-bipyridylium ion)

+ +

N N

DQ (1,1′-ethylene-2,2′-dipyridylium ion)

Fig. 1 – Chemical structures of paraquat (PQ) and diquat (DQ).

PQ was first described in 1882 (Weidel and Rosso, 1882). Its redox properties

were discovered in 1933 by Michaelis and Hill (Michaelis and Hill, 1933). By that time

it was used as an oxidation-reduction indicator because an electron donation to the PQ

ion (PQ 2+ ) forms a stable free radical monocation (PQ •+ ) having a violet or blue colour

(Michaelis and Hill, 1933); hence, PQ is commonly called methyl viologen (Fig. 1).

Part I - General Introduction __________________________________________________

4

The PQ herbicidal properties were discovered at the Jealott’s Hill International

Research Centre, Bracknell, UK in 1955, and in August 1962, PQ was introduced into

the market as an herbicide by the Plant Protection Division Ltd of Imperial Chemical

Industries [(ICI), now Syngenta] (Homer et al., 1960; Calderbank, 1968; Smith and

Heath, 1976). Gramoxone®, manufactured by Syngenta, is the most common trade

name for PQ, but the herbicide is also sold under many different trade names by several

different companies (Table 1).

Table 1 – Some paraquat trade names.

Paraquat Paraquat-Diquat Mixtures

Crisquat Preeglone

Dextrone X Priglone

Esgram Weedol

Gramoxone

In spite of the numerous intoxications, PQ is now registered and used in over 120

developed and developing countries throughout the world (Table 2). The main reasons

for such widespread use are the following:

• PQ is an excellent herbicide for destroying weeds that may decrease crop

yields, during normal application. It is also used in pasture renovation and on

non-crop areas such as public airports, electronic transformer stations, and

around commercial buildings. Its success as a weed-killer lies on the fact that

small quantities of a PQ solution will rapidly kill plants on contact with the

leaves and due to its low cost. In addition, PQ allows the roots to remain intact,

thus holding the soil together and preventing soil erosion;

• PQ is highly hydrophilic and thus not absorbed through intact skin;

• Aerosolized PQ particles are larger than 5 μm in diameter and thus do not reach

the humans alveoli when exposed by inhalation route;

• PQ is rapidly inactivated and metabolized once in the soil, preventing its

accumulation in the ecosphere.

__________________________________________________Part I - General Introduction

Table 2 - Countries in which paraquat is registered as of June 2005. Adapted from

paraquat information center (www.paraquat.com).

Albania El Salvador Malaysia Sierra Leone

Algeria Ethiopia Mali Singapore

Angola Fiji Malta Slovakia

Antigua & Barbuda France Mauritania Somalia

Argentina Gabon Mauritius South Africa

Australia Gambia Mexico South Korea

Bahamas Germany Morocco Spain

Bahrain Ghana Mozambique Sri Lanka

Bangladesh Greece Myanmar Sudan

Barbados Grenada Namibia Suriname

Belgium Guatemala Netherlands Swaziland

Belize Guinea New Zealand Tahiti

Bolivia Guinea-Bissau Nicaragua Taiwan

Botswana Guyana Niger Tanzania

Brazil Haiti Nigeria Thailand

Burkina Faso Honduras Oman Trinidad & Tobago

Burundi India Pakistan Turkey

Cameroon Indonesia Panama Uganda

Canada Iran Papua New Guinea United Kingdom

Cape Verde Iraq Paraguay USA

Chad Ireland Peru Uruguay

Chile Israel Philippines Venezuela

China Italy Poland Vietnam

Colombia Jamaica Portugal Yemen

Costa Rica Japan Romania Yugoslavia

Cote d’Ivoire Jordan Rwanda Zambia

Croatia Kenya Sao Tome & Principe Zimbabwe

Cuba Lebanon St Kitts & Nevis

Czech Republic Liberia St Lucia

Dominica Macedonia St Vincent & Grenadines

Dominican Republic Madagascar Samoa

Ecuador Malawi Senegal

Since its introduction in the market, numerous successful practical uses of the

herbicide have been implemented. PQ is an extremely effective, fast-acting and non-

selective foliage-applied contact herbicide, killing a wide range of grass and dicot

weeds. It is a defoliant, desiccant, and plant growth regulator herbicide. PQ is rapidly

inactivated by the majority of surrounding soils in the event of overspray (Amondham

et al., 2006). Inactivation on contact with soil means that no biologically active residues

remain in the soil, thus allowing planting or sowing to be carried out almost

immediately after spraying. The PQ 2+ is strongly attracted to the negative charge of soil

clay particles and once the equilibrium is established (Fig. 2), PQ, at typical

Part I - General Introduction __________________________________________________

environmentally expected concentrations, becomes a strongly adsorbed residue that is

biologically unavailable due to having an extremely low concentration in the soil

solution. The soil’s natural deactivation capacity for this herbicide is several times the

normally recommended application rate (Smith and Oehme, 1991), existing evidences

demonstrating that adsorption is capable of deactivating the equivalent of hundreds or

even thousands of PQ applications over a wide range of soils. This also means that PQ

is effectively immobilized in soils with no leaching to ground water (Roberts et al.,

2002).

6

Soil Surfaces Soil Pore Water

>99.99% Microbial

Strongly Adsorbed

Degradation

CO 2 + H 2 O

Fig. 2 – Equilibrium dynamics for paraquat between the soil and soil solution. Adapted

from Roberts et al. (Roberts et al., 2002).

Although the non-systemic (contact) property of PQ makes it less than ideal for

the long-term control of perennial weeds, the same property is of real advantage when

parts of crop plants are sprayed accidentally and thus only the part receiving the spray is

affected (Sagar, 1987). PQ is only rainfast within minutes of application. This property

reduces the operator’s dependence on weather and allows great precision in the timing

of applications. In relation to the crop, PQ may be applied preharvest, preemergence, or

preplant. The herbicidal activity becomes obvious as a rapid decolourization and

desiccation of green plant tissue when illuminated. PQ has also been used for control of

aquatic weeds in irrigation ditches from where residues disappear rapidly due to its

strong adsorption to bottom mud and onto aquatic weeds (Grover et al., 1980). The best

herbicidal results are achieved by spraying in late afternoon or evening. However, the

herbicidal activity is slower in the dark, owing to the absence of naturally occurring

reducing agents during photosynthesis.

Product formulations differ among countries. Typically, it is available as a 10% to

30% concentrated solution (according to the manufacturer’s instructions, correctly

diluted spray solutions should contain no more than 0.05 to 0.2% of PQ ion) for

agricultural use or as a 2.5 or 5% powder (w/w) for domestic use. Granular and gel

forms are also encountered. PQ is caustic, and the concentrate may also contain an

__________________________________________________Part I - General Introduction

aliphatic detergent to enhance entry of PQ into the cells and thus its toxicity. PQ can be

applied safely when used according to the manufacturer’s guidelines (Hart, 1987). It is

generally with the liquid formulations (specially the concentrated ones) for agricultural

use that the vast majority of fatal cases of intoxication have occurred. Proudfoot et al.

(Proudfoot et al., 1987) reported a mortality of 65% in patients who ingested the

concentrated formulation and only 4% in those who ingested the diluted solutions (2.5%

w/v). In the granular form, it is difficult to ingest by accident, and large quantities of the

granules have to be ingested to induce toxic effects. Originally, marketed aqueous PQ

formulations were brown in colour. Due to mistakes with other common beverages such

as coffee, cola drinks, among others, the colour is now dark blue-green. The

formulations also contain a powerful stenching and emetic agent. Recently, Syngenta

scientists have developed a formulation, Gramoxone Inteon®, which contains a gelling

agent (alginate) that is activated or triggered at the pH of stomach acid, increased levels

of emetic and a purgative. Once formed, the gel minimizes and slows dispersion and

passage of PQ into the small intestines, thus allowing more time for productive emesis

as caused by the emetic. The new formulation improved overall survival following PQ

ingestion from 25.6% to 35.3% (www.paraquat.com).

1.1 Mode of action as herbicide

Herbicidal activity as well PQ-induced toxicity to mammals was found to be

linked to PQ redox potential (Bird and Kuhn, 1981). Initial work on the mode of action

of BHs by Mees (Mees, 1960) indicated that their ability to cause rapid kill is dependent

on the photosynthetic activity of plants and on oxygen (O2). Zweig et al. (Zweig et al.,

1965) further found that BHs cause a deviation of electron flow from Photosystem I

(which normally transfers its electron to ferredoxin), leading to an inhibition of oxidized

nicotinamide adenine dinucleotide phosphate (NADP + ) reduction during photosynthesis

(Fig. 3). As result of this process, a PQ •+ is produced in the cell at the expense of

NADPH. Thus, PQ is only toxic to the green parts of the plant, where the

photosynthesis occurs (Slade, 1966). When plants are irradiated with sunlight, the

generated electrons reduce PQ 2+ to PQ •+ , which is rapidly reoxidized by the O2

produced in chloroplasts (Slade, 1966). During the reoxidization a superoxide radical

(O2 .- ) is generated, with the subsequent deleterious effects and consequent cell death.

Part I - General Introduction __________________________________________________

Therefore the mechanism of toxic action of PQ involves cyclic reduction–oxidation

reactions, which produce reactive oxygen species (ROS) and depletion of reduced

nicotinamide adenine dinucleotide phosphate (NADPH). An excellent review of the

subject is given by Dodge (Dodge, 1971).

Fig. 3 – Herbicidal mechanism of paraquat. In photosystem I (PS I), plastocyanin (PC)

transfers its electron (e - ) through a series of steps (P700, A0, A1, FX, FA) to ferrodoxin

(FRD) and finally to NADP + . PQ ion (PQ 2+ ) binds near the FRD binding site in PS I and

accepts an e - , becoming paraquat monocation free radical (PQ •+ ), which initiates a series

of reactions leading to cell membrane disruption and plant death. The formation of such

free radicals stops electron transport to NADP + and effectively inhibits normal

functioning of PS I. A, Ferrodoxin-NADP + reductase.

1.2 Biodegradation Pathways

When bound to soil, PQ is biologically inactive, unavailable for either herbicidal

or ecotoxicological action and unavailable for microbial degradation or

8

Stroma

Lumen

PQ 2+

PsaD

hv

PsaA

FA FB

e- e- e- e- e- e- e- e- PC 2+

F X

A 1

A 0

P 700

PC +

NADP + + H +

A

X

FRD ox FRD red

e- e- PsaB

PsaF

PsaE

PsaC

PQ .+

PSI

NADPH

O 2

O 2 .-

PQ 2+

CELLULAR

DAMAGE

__________________________________________________Part I - General Introduction

photodecomposition (Burns and Audus, 1970; Smith and Oehme, 1991). The strong

binding of PQ to clay minerals (e.g., bentonite) forms the basis of a suggested method

for preventing systemic absorption in human poisoning cases (Smith et al., 1974a).

Such soil-bound residues may persist essentially indefinitely, with only a 10% annual

loss and a field half-life (t1/2) of 6.6 years (Hance et al., 1980). PQ is only significantly

available for degradation during the immediate period after soil application (especially

during the first 96 hours), when the herbicide is only weakly adsorbed to the soil

particles (Burns and Audus, 1970).

1.2.1 Photochemical Degradation

The photochemical degradation of PQ has been observed in laboratory conditions,

as well as on the surface of plant matter and soils. It is the predominant mechanism of

PQ degradation in soils (Smith and Mayfield, 1978) and it is related to the availability

of UV between the wavelengths of 290 and 310 nm during daylight hours (Slade, 1965;

Slade, 1966). The main intermediates of photochemical PQ degradation on plants or soil

surfaces are of low toxicity. They decompose easily and are not expected to produce

adverse environmental effects.

1.2.1.1 On plant surfaces

In agricultural practice, much of the sprayed PQ is initially deposited on plant

surfaces. Slade (Slade, 1965; Slade, 1966) applied PQ dichloride droplets to maize,

tomato, and broad-bean plants and studied the degradation pathways. Determinations

carried out at intervals of 100 days showed that degradation was caused by

photochemical decomposition on the leaf surfaces and not by metabolism. Degradation

products isolated from plants sprayed with [ 14 C]PQ dichloride included 4-carboxyl-1methyl-

14 C-pyridylium chloride or 4-methylisonicotinic acid (MINA) and methylamine-

14

C-hydrochloride. The photochemical degradation of PQ dichloride continued after the

plants were dead (Fig. 4). The photochemical degradation of PQ is rapid. A 0.1% PQ

solution was completely degraded in 3 days under a UV lamp (Slade, 1965). PQ

photodegradation products were not translocated from the desiccated leaves of the

H3C

O

N

Part I - General Introduction __________________________________________________

plants, nor were they found in the crops (cereals and fruits), when weeds were treated

with PQ during 3-4 successive seasons (Slade, 1965).

10

O

N

CH 3

C

H 3

C

H 3

O 2 , UV radiation

+ +

N

N CH3 + +

N

C N

PQ monopyridone

O

O 2 , UV radiation

C

H 3

CH 3

+

N

microbial

H 3

NH 3 + CO 2 + H 2 O

O 2 , UV radiation, microbial

COOH

N-methylisonicotinic acid (MINA)

Monoquat

O 2 , UV radiation, microbial

CO 2 + CH 3 NH 2 HCl + formate + oxalate + succinate

O 2 , UV radiation

Fig. 4 – Photochemical and microbial degradation of PQ. Adapted from Slade (Slade,

1965).

1.2.1.2 On soil and other mineral surfaces

Slade (Slade, 1966) showed that there was a breakdown, similar to that observed

on plant surfaces, if spots of PQ on silica gel were directly exposed to sunlight. When

[ 14 C]PQ dichloride was sprayed on the bare soil surface of a field during a hot sunny

period, traces of MINA were detected in the top inch of soil for the first few weeks

afterwards (Calderbank and Slade, 1976). Radioassay showed that the total soil residue

did not markedly decrease during a 6-18 month period, so that, in agricultural practice,

UV degradation of herbicide reaching the soil should be regarded as insignificant.

N

CO 2 + CH 3 NH 2 HCl

__________________________________________________Part I - General Introduction

1.2.2 Microbial degradation

Microbial PQ degradation has been thoroughly reviewed by Haley (Haley, 1979).

In soils, it does not occur at appreciable rates due to the sequestering of the herbicide at

mineral or organic anionic sites. Nevertheless, the biodegradation of PQ has been

observed by a wide variety of soil microorganisms in aqueous solution. Baldwin et al.

(Baldwin et al., 1966) identified many soil microorganisms capable of degrading PQ.

The herbicide was decomposed by Corynebacterium fascians, Clostridium

pasteurianum, and Lipomyces starkeyi. Several other microorganisms were found to

degrade PQ (Smith and Heath, 1976) but Lipomyces starkeyi proved to be the most

active (Burns and Audus, 1970). In pure culture, degradation of PQ has been reported to

occur under both aerobic and anaerobic conditions by Clostridium pasteurianum

(Baldwin et al., 1966). In the case of the yeast Lipomyces starkeyi, the ability to degrade

PQ was only seen under aerobic conditions (Funderburk, 1969).

Several decomposition products of PQ have been characterized. The demethylated

product, 1-methyl,-4,4’-bipyridylium (Fig. 4) was recovered from an unidentified

bacterial culture as well as the N-methyl betaine of isonicotinic acid or MINA

(Summers, 1980). Whether these compounds occur sequentially in one degradation

pathway or in separate pathways is unknown. The N-methyl betaine has been shown to

be readily degraded in soils into methylamine and CO2 by microbial activity (Wright

and Cain, 1970). The pyridylium ring carbons are known to be lost as CO2 by 14 C

labelling studies. The methylamine can be utilized as a source of nitrogen and carbon.

However, the enzymology of the degradation of PQ has not been reported and the other

intermediates have not been identified.

2. CHEMISTRY OF PARAQUAT

The chemistry of PQ is dominated by its ability to act as a one-electron carrier.

The electron can be transferred to PQ either partially (from a nucleophile or other

electron-rich compound) or completely (from a reducing agent). In the first case,

coloured charge-transfer complexes are formed; in the second case, the blue-coloured

PQ •+ is formed.

Part I - General Introduction __________________________________________________

2.1 Physical and chemical properties

12

The physical and chemical properties of PQ 2+ are summarized in Table 3. PQ is

highly water soluble, slightly soluble in alcohol and practically insoluble in organic

solvents (Haley, 1979). PQ is non-explosive and non-flammable in aqueous

formulations. It is corrosive to metals and incompatible with alkylarylsulfonate wetting

agents and strong oxidizing substances. Although non-ionic surfactants may be used in

combination, PQ is inactivated by anionic ones. It is stable in acid or neutral solutions

but is readily hydrolysed by alkaline solutions (at pH >12). In the original container and

under normal conditions, the shelf life of PQ is indefinitely long and it is also stable at

temperatures above the general environmental range.

Table 3 - Physical and chemical properties of PQ ion. 1 1 g of paraquat dichloride =

0.724 g of paraquat ion.

Class bipyridylium herbicide

Molecular formula C12H14N2

Molecular weight 186.3 (ion), 257.2 (dichloride) 1

Common name paraquat

IUPAC name 1,1-dimethyl-4,4-bipyridilium

CAS name 1,1-dimethyl-4,4-bipyridilium

Synonyms methyl viologen

CAS Nº

4685-14-7 (ion), 1910-42-5 (dichloride) and

2074-50-2 (sulphate)

Specific gravity (20ºC) 1.240-1.260 g/cm 3

Physical state

Melting point

Boiling point

Solubility in water at 20°C 700 g/L

pH of liquid formulation 6.5-7-5

white (pure salts), yellow (technical products)

crystalline, odorless, hygroscopic powders

PQ dichloride melts with decomposition at

~340ºC to form poisonous vapors

PQ dichloride decomposes at ~340ºC to form

poisonous vapors

__________________________________________________Part I - General Introduction

Vapor pressure not measurable

E0’ (relative to the normal

hydrogen electrode

Octanol/water partition

coefficient as log Pow (20ºC)

– 0.446 V

-4.2

Dissociation constant PQ ion does not dissociate

Relative gas density 8.88

UV spectrum (in water)

X-ray analysis of the crystalline

dichloride

Single band centered at 257 nm (Kosower and

Cotter, 1964)

Show two coplanar pyridine rings with two

methyl groups (Russell and Wallwork, 1972)

The basic chemical nucleus of PQ (Fig. 1) is a bipyridylium consisting of two

quaternized pyridine rings bonded together such that their nitrogen atoms face

diametrically away from one another. The quaternization is the result of the methyl

radical addition (para position) to each of two nitrogen nuclei in the pyridine rings. The

compound is, therefore, a para substituted quaternary bipyridylium, hence its common

designation PQ. In its usual oxidized form, it is ionized (bearing two positive charges)

and it is most commonly manufactured as a dichloride salt. Chemically, PQ is thus 1,l’dimethyl-4,4’

dipyridylium dichloride. The positive charges are largely resident on the

nitrogen atoms as shown by nuclear magnetic resonance (NMR) (Smith and Schneider,

1961). The herbicidal efficacy of PQ is related to the concentration of free PQ 2+ in

solution inside the chloroplast, but this will depend (in some cases, markedly) on the

nature of the anion and any complexing agent with which it is applied (Homer and

Tomlinson, 1959). For instance, many phenols form soluble crystalline complexes with

PQ dichloride (White, 1969; Ledwith and Woods, 1970). This ready formation of

complexes is possibly one of the factors that change the herbicidal activity of PQ among

floral species and through the plant lifespan. Many plants constituents, such lignin and

tannin, are phenolic in nature and could cause immobilization of PQ (White, 1969;

Ledwith and Woods, 1970).

2.2 Synthesis

PQ does not occur naturally. It was originally synthesized by Weidel and Rosso as

reported in 1882 (Weidel and Rosso, 1882). There are several methods available for the

Part I - General Introduction __________________________________________________

synthesis of PQ. In the most common method, PQ is produced by coupling pyridine in

the presence of sodium in anhydrous ammonia and quaternizing the 4,4’-bipyridyl with

an excess of methyl chloride to obtain PQ dichloride (Fig. 5). When bipyridyl is

refluxed with methyl iodide or methyl bromide, the iodide and the bromide salt is

obtained, respectively. The methyl sulfate salt can be obtained by heating 4,4’-bipyridyl

with sodium acetate at 70ºC for 2 hours, then adding methyl sulfate and stirring for 15

min. Haley and Summers (Haley, 1979; Summers, 1980) thoroughly reviewed the

published methods for PQ synthesis, and for the separation and purification of

bipyridylium salts. The yields obtainable vary from 20% to 96% of pure product. The

only impurity permitted in the final product is the 4,4’-bipyridyl at a maximum level of

0.25% of the PQ content (Summers, 1980).

14

2Cl -

+

2 N

N

C

H 3

Fig. 5 - Synthesis of paraquat.

+

N

+

2CH 3 Cl

Na/NH 3 /O 2

N

+

N CH3

C

H 3

__________________________________________________Part I - General Introduction

2.3 Electrochemistry of viologens and paraquat reduction

The viologens exist in three main oxidation states, namely V 2+ ↔ V •+ → V. The

first reduction step is highly reversible and can be cycled many times without

significant side reaction. The further reduction to the fully reduced state is less

reversible, not only because the latter is frequently insoluble but also because it is an

uncharged one. The compounds are also very stable chemically, although in more

alkaline solutions they will dealkylate (Fig. 6) as reported by Farrington et al.

(Farrington et al., 1969). Because the methanol resulting from the dealkylation can be a

reducing agent, solutions of methyl viologen in alkali can spontaneously be reduced and

will then turn blue as the PQ •+ is formed. Methanol is then itself oxidized to

formaldehyde.

OH - CH 3 OH

+ +

+

N

N CH N

N CH 3

3

Fig. 6 - Paraquat dealkylation in alkaline solutions.

The PQ divalent cation is colorless, whereas the partially reduced PQ •+ is blue

colored and contains an odd electron. The odd electron is shared by all the nuclear

carbon positions in the rings (Calderbank, 1968). This step is completely reversible,

such that one equivalent of a reducing agent will reduce more than 50% of PQ 2+ to PQ •+

only if its reduction potential is more negative than that of PQ. Ito and Kuwana (Ito and

Kuwana, 1971) quoted the potential of the first reduction for PQ 2+ as -0.446 V and the

second is given as -0.88 V (relative to normal hydrogen electrode). A suitable reducing

agent for generating the radical is sodium dithionite in alkaline solution (-1.13 V).

Considering that PQ •+ carries an unpaired electron in a π anti-bonding orbital, PQ •+ is

remarkably unreactive. In addition, unpaired electron in PQ •+ is not fixed.

Delocalization of the unpaired electron gives considerable resonance stabilization to the

radical and thus it may diffuse outside the cell before reacting with O2 (Fig. 7). It is a

known fact that the greater the number of positive centres available to an odd electron,

the greater the number of resonance structures and the higher the stability of the free

Part I - General Introduction __________________________________________________

radical. For review see (Akhavein and Linscott, 1968). The second step of PQ reduction

is the addition of a second electron to the molecule to originate the 1,l’-dimethyl-4,4’dihydrobipyridyl.

If it is required to stop the reduction at the radical stage, either a

limited amount of the reducing agent must be used or the solution must be “poised” to

the desired reducing potential (-0.56 V for 99% conversion to radical) by adjusting the

concentration of the reductant or the pH if relevant. The resulting fully reduced species

is colourless (Michaelis and Hill, 1933).

16

C

H 3

C

H 3

C

H 3

C

H 3

+

N

+ +

N

.

N

- e -

.

...

+ e -

+

N CH3 +

N CH3 N

CH 3

1,1'-dimethyl-4,4'-dihydrobipyridyl

Fig. 7 – Intermediate resonance structures of paraquat and full reduction.

__________________________________________________Part I - General Introduction

3. TOXICOKINETICS OF PARAQUAT

The toxicokinetics of PQ has been studied in a variety of animal species,

especially dogs, rats, and rabbits (Murray and Gibson, 1972; Hawksworth et al., 1981;

Yonemitsu, 1986). The dog seems to be the most similar model of PQ to human

toxicokinetics (Hawksworth et al., 1981).

3.1 Absorption

Nearly all PQ poisonings result from ingestion. PQ is known to be very rapidly

absorbed, apparently associated with the carrier-mediated transport system for choline

on the brush-border membrane, thought this absorption from the gastrointestinal tract

(GIT) is low (Nagao et al., 1993a). Absorption occurs primarily in the small intestine

(poorly from the stomach) and is estimated to be 1-5% in humans over a 1–6 hours

period (Baselt and Cravey, 1989; Houze et al., 1990; Houze et al., 1995). Any recent

food ingestion may decrease the amount of systemic absorption (Meredith and Vale,

1987; Bismuth et al., 1988). Although the plasma peak time (Tmax) is not known with

certainty in humans, PQ may be detected in the urine as early as 1 hour after ingestion,

and according to data published by Proudfoot et al. (Proudfoot et al., 1979; Proudfoot,

1995) and Smith (Smith, 1988b), peak concentrations (Cmax) in humans are attained

within 4-and possibly within 2-hours after intoxication. Smith (Smith et al., 1974a)

reported that after oral administration of PQ to rats, plasma concentrations remained

relatively constant for 30 hours. During this period of time, concentrations in the lung

rose progressively to several times the plasma concentration. If during the first 30 hours,

plasma PQ concentrations were severely reduced by decreasing absorption of the

herbicide from GIT or increasing its elimination by extracorporeal techniques from the

plasma, lethal concentrations woudn’t reach the lungs (Smith et al., 1974a). These

authors also concluded that not only the Cmax is responsible for determining the lung

levels but also the maintenance of plasma levels from which the lung can take large

amounts of PQ. The maintenance of such plasma concentrations in the rat has been

shown to be the result of continued PQ absorption from the GIT over the first 30 hours

after oral administration. Absorption of PQ from the GIT into the human bloodstream is

Part I - General Introduction __________________________________________________

quite different from that seen in rats; concentrations declined rapidly over the first 15

hours after Tmax to much lower levels than those described in rats due to tissue

distribution, and more slowly thereafter (Smith, 1987). Thus, in humans, if adsorbents

are to be effective in preventing PQ from entering the blood and consequently the lung,

they must be administered within a few hours, or accordingly to Bismuth et al.

(Bismuth et al., 1988), within the first few minutes after ingestion.

18

Daniel and Gage (Daniel and Gage, 1966) studied the absorption of [ 14 C]-PQ

following oral and subcutaneous (s.c.) single-dose administration to rats. About 76-90%

of the oral doses were found in the faeces, and 11-20% in the urine; most of the s.c.

dose (73-88%) was found in the urine and only 2-14.2% in the faeces. These values bear

no relation to their respective LD50 values (Conning et al., 1969). These studies

evidenced that PQ was poorly absorbed from the gut. Rats, guinea-pigs, and monkeys

orally administered LD50 doses of [ 14 C]-PQ had low peak plasma concentrations (2.1-

4.8 mg/L) (Murray and Gibson, 1972). Extensive caustic injury to the GIT may increase

the amount absorbed. The highest concentration was found 1 to 6 hours after an oral

dose depending upon the species used (Murray and Gibson, 1972).

Although almost all fatal exposures have resulted from the ingestion of PQ, a few

case reports have involved rather extensive skin contamination (Samman and Johnston,

1969; Hearn and Keir, 1971; Vale et al., 1987; Smith, 1988a; Hoffer and Taitelman,

1989). PQ absorption through animal and human skin was studied in vitro (Walker et

al., 1983). Human skin was shown to be impermeable to PQ, having a very low

permeability coefficient of 0.73. Furthermore, human skin was found to be at least 40

times less permeable than the animal skins tested (including rat, rabbit, and guinea-pig)

(Walker et al., 1983). A study of the percutaneous absorption of PQ was undertaken in

six human volunteers by Wester et al. (Wester et al., 1984). It was observed that only

minute quantities of PQ were absorbed through intact human skin over 24 hours and

that there was little difference among skin tested at different body sites in its ability to

absorb PQ.

Fatal cases of s.c., intravenous (i.v.), intramuscular or intraperitoneal (i.p) injection

of PQ were also reported, the doses being considerably lower than the lethal dose by

ingestion route (Almog and Tal, 1967; Vale et al., 1987; Hsu et al., 2003).

Ocular exposure may cause local caustic injury with ulceration and scarring likely

resulting in a delayed slough of corneal epithelium 12–24 hours after exposure, but not

resulting in systemic toxicity (Cant and Lewis, 1968; McKeag et al., 2002).

__________________________________________________Part I - General Introduction

Inhalation of PQ used in an agricultural/occupational setting does not allow

sufficient absorption to cause systemic disease, because of droplet size (greater than 5

µm) that prevents deep lung exposure and absorption, low product vapour pressure, and

low application concentration (Howard, 1983; Chester and Ward, 1984).

Notwithstanding no fatal cases have been reported from inhalation of PQ vapor or

aerosols, toxicity has occurred from this route of exposure, since inhalation of PQ

droplets may produce nasal and tracheobronchial irritation. An interesting episode in the

history of the war against illicit marijuana use, in which large quantities of PQ were

sprayed over culture fields in the early 1970s is described. By then, PQ was the

herbicide of choice during aerial spraying of marijuana by the U.S.A. and Mexican

governments. However, after spraying, growers simply harvested the crops before the

plants were exposed to enough sunlight to damage the plants, resulting in an apparently

healthy harvest though contaminated with PQ. Concerns regarding the smoking of PQ-

sprayed marijuana in the early 1970s has proved unfounded, because PQ is destroyed by

pyrolysis into a relatively nontoxic compound (4,4’-bipyridyl) during the smoking

process (Groce and Kimbrough, 1982; Landrigan et al., 1983).

A fatal case of PQ absorbed per vagina of a 28-year-old woman (who inserted a

tampon inadvertently soaked in PQ) as a consequence of respiratory, renal and hepatic

dysfunction was reported (Ong and Glew, 1989).

3.2 Distribution

Despite numerous studies, the distribution of PQ through the different tissues is

still unclear. Dey et al. (Dey et al., 1990) studied the toxicokinetics of [ 14 C]-PQ in rats

exposed to a single s.c. injection. PQ was rapidly absorbed with a Tmax of 20 min. The

toxicokinetic was best characterized by a two-compartment open model, the mean t1/2

being approximately 40 hours. Peak concentrations in the kidney and lung tissues were

at around 40 min. Nevertheless, the majority of the authors agree that the kinetics of PQ

in the plasma is better described by a three-compartment open model. Murray and

Gibson described a triexponential disappearance of [ 14 C]-PQ from the plasma after oral

administration (126 mg/Kg) in rats, guinea pigs and monkeys (Murray and Gibson,

1972). The toxicokinetics of PQ appears to be similar in human and dog (Hawksworth

et al., 1981; Vandenbogaerde et al., 1984). Hawksworth et al. (Hawksworth et al.,

Part I - General Introduction __________________________________________________

1981) described a plasma-concentration-time curve with a triexponential decline, in

dogs, suggesting a three-compartment model:

-Blood is assumed to be the central compartment. The concentrations found in

plasma and erythrocytes are approximately the same at least in the rat (Sharp et al.,

1972);

-The shallow compartment is though to be composed of highly perfused tissues

such as the kidney, liver, heart, etc. Rapid exchanges occur between this compartment

and blood. The anatomy and physiology of the lung (a highly vascularized tissue)

suggests therefore early exposure to any PQ circulating in the blood (Bismuth et al.,

1987);

-The third compartment lies within the lungs, especially the pneumocytes type I

and II and Clara cells where exchanges with the central compartment are slow (Bismuth

et al., 1987). The initial t1/2 of PQ in the lung was much greater than the t1/2 in other

vital organs (e.g. kidney, liver, muscle, adrenal, spleen, heart, testis), explaining the

highest PQ lung accumulation (Sharp et al., 1972). The kinetics of PQ in the rat lungs

shows a rapid decline with an elimination t1/2 of 20 min, followed a slower decline with

a t1/2 of about 50 hours (Sharp et al., 1972). Peak concentration in lungs is reached 4-5

hours after i.v. administration, and 5-7 hours after ingestion, provided that renal

function is normal (Sharp et al., 1972). Lethal concentrations may be achieved in the

lung within 6 hours of ingestion of 35 mg/Kg (Houze et al., 1995). Patients are only

rarely admitted to an experienced hospital in the treatment of PQ poisonings before the

pulmonary peak. In the presence of renal failure (which normally occurs when more

than 20 mg/Kg PQ are ingested), peak pulmonary concentration is not achieved for 15-

20 hours, and may reach very high values (120 hours or longer) (Hawksworth et al.,

1981; Bismuth et al., 1987). Renal failure precludes elimination of PQ by its normal

route. Furthermore, Hawksworth et al. (Hawksworth et al., 1981) suggested that an

impairment of renal function by as little as 5% produces a five-fold higher concentration

of the herbicide in the plasma. A critical plasma threshold is needed for active

pulmonary uptake to occur (Manabe and Ogata, 1987). With time, however, it was

shown that the concentration in the lung did fall to below that in muscle (due to the

secondary t1/2 in the muscle). Considering that muscle represents a large percentage of

the body mass, it may be considered an important reservoir of PQ (Murray and Gibson,

1972; Sharp et al., 1972).

__________________________________________________Part I - General Introduction

Houze et al. (Houze et al., 1990) studied the toxicokinetics of PQ in 18 cases of

acute human poisoning. The concentration-time course was described by using a

biexponential curve suggesting a two-compartment model with absorption, distribution

and elimination phases. Plasma PQ concentration exhibited a mean distribution half-life

(t1/2α) of 5 hours and a mean elimination half-life (t1/2β) of 84 hours. Tissue PQ

distribution was ubiquitous with an apparent volume of distribution ranging from 1.2 to

1.6 L/Kg. Muscle represented an important reservoir explaining the long persistence of

PQ in plasma and urine for several weeks or months after poisoning (Smith, 1988b).

The volume of distribution of PQ estimated from a kinetic study in one patient was 2.75

L/Kg (Davies, 1987). Recently, immunohistochemical studies were used to demonstrate

the distribution and localization of PQ in several organs. In the skin, PQ was localized

in the ducts of sweat glands and sebaceous glands between 3 and 10 days after PQ i.v.

injection (Nagao et al., 1993c). In the eyes, weak positive findings were observed in

nerve fibers of retina between 3 and 10 days after the injection. In the cornea, PQ was

localized in epithelial cells at the first 3 hours and between 3 and 10 days after PQ

administration. Since skin occupies a vast area of the body in animals, as an organ, it

seems to be an important storage pool for the redistribution of PQ (Nagao et al., 1993c).

PQ was also found in immune and haematopoietic systems (Nagao et al., 1994). In the

bone marrow, PQ was localized in several types of blood cells (granulocyte, erythrocyte

and megakaryocyte) and their precursors between 24 hours and 7 days after the i.v.

administration. In the thymus, PQ was mainly localized in the medulla between 12

hours and 7 days after administration, whereas in the spleen, it was mainly localized in

the red pulp between 12 hours and 10 days after administration of PQ (Nagao et al.,

1994). In the stomach, PQ was localized in the epithelial cells between 24 hours and 10

days after administration, whereas in the esophagus, PQ was localized in epithelial cells

and the lamina propria mucosa between 12 hours and 10 days after administration.

Although these findings were similar to those observed in the intestine of rats, no clear

changes in the distribution of PQ with time were observed, suggesting that the stomach

and esophagus are important reservoirs for the redistribution of PQ (Nagao et al.,

1993b).

Concerning the PQ binding to plasma proteins, controversial data exist. For many

years PQ was thought not to bind to plasma proteins (Lock and Ishmael, 1979).

Recently Jaiswal et al. (Jaiswal et al., 2002) showed the binding of PQ to plasma

albumin by using a fluorescence technique.

Part I - General Introduction __________________________________________________

3.2.1 Preferential accumulation in the lung

22

Irrespective of the route of administration, the lung and the kidney are the organs

showing the highest concentrations of PQ (Murray and Gibson, 1972; Sharp et al.,

1972; Ilett et al., 1974). The distribution of [ 14 C]PQ into various tissues after oral

administration of 680 mol/Kg to rats was followed as a function of time by Rose et al.

(Rose et al., 1976a). They showed that although the plasma concentration of PQ

remained constant between 2 and 30 hours after administration, PQ concentrations in

the lung exhibited a time-dependent increase over the same period. None of the other

studied organs showed this time-dependent accumulation. Rose et al. (Rose et al., 1974)

also demonstrated that slices of lung incubated with [ 14 C]PQ exhibited a time-dependent

accumulation of radioactivity. In addition, lung slices were the only tissue slices in

which PQ accumulated at a concentration significantly higher than that in the medium.

These studies demonstrated that lung, and no other major tissue, is able to accumulate

PQ against a concentration gradient. After in vivo administration, PQ levels in the

kidney did not show a time-dependent increase, but were nevertheless higher than those

in the lung throughout the first 30 hours (Rose et al., 1976a). These high concentrations

of PQ in the kidney probably result from the fact that this organ represents the

predominant route of elimination of PQ from the circulation and are likely to constitute

extracellular rather than intracellular PQ. They may also underlie the observation that

renal failure often occurs in PQ poisoning, especially during early stages. Taken

together, these studies strongly suggest that the lungs are a specific target for the

pathological effects of PQ because of its selective accumulation by this organ. PQ

pulmonary concentrations can be 6 to 10 times higher than those in the plasma, and the

compound is retained in the lung even when blood levels are starting to decrease.

3.2.2 Lung accumulation through the polyamine uptake system

Early work by Rose et al. (Rose et al., 1974) demonstrated that the accumulation

of PQ into rat lung slices occurred against a concentration gradient and could be

abolished by metabolic inhibitors such as cyanide or rotenone, suggesting that the

uptake is an adenosine triphosphate (ATP)-driven process. The accumulation also

exhibited saturation kinetics with an apparent Michaelis-Menten constant (Km) of 70 µM

__________________________________________________Part I - General Introduction

and a maximal rate (Vmax) of 300 nmol PQ × g wet weight 1 × h 1 . These observations,

coupled with findings that PQ is neither metabolized by the lung (Conning et al., 1969;

Ilett et al., 1974) nor becomes covalently bound to any degree (Ilett et al., 1974;

Sullivan and Montgomery, 1983), suggest that its accumulation is mediated through

binding to, and subsequent translocation into cells by a carrier system. Active

accumulation of PQ via transport systems exhibiting similar kinetic parameters (Table

4) was also demonstrated in lung slices taken from other species [beagle dogs, New

Zealand white rabbits, and cynomolgus monkeys (Macaca fascicularis)], including

humans (Rose et al., 1976a). The kinetic constants for human and rat lung are

statistically similar, suggesting that the rat may be a good experimental model for the

study of PQ accumulation in the human lung. Thus, it seems likely that the human lung

does possess a similar transport to that characterized in the rat lung and this process

accounts for the selective accumulation and hence the selective toxicity of PQ to the

human lung.

Table 4 - Kinetic constants for the accumulation of paraquat into lung tissue slices from

various species Adapted from Rose et al. (Rose et al., 1974).

Species Km (µM)

Vmax

(nmol of PQ/g

tissue/hour)

Rat 70 300

Mouse 68 556

Syrian hamster 77 452

Guinea-pig 96 49

Rabbit 0.05 20

Man 40 300

Monkey 70 50

Dog 60 10

Carrier-mediate PQ uptake also occurs in the isolated perfused lung, into which

accumulation of PQ to a concentration in excess of that in the perfusate has been

observed (Rannels et al., 1985). However, the kinetics of this process appears somewhat

different in comparison to the lung slices. The onset of active transport is preceded by

an initial lag phase during which the intracellular PQ concentration approaches that in

the perfusate, possibly because the endothelium functions as a barrier between the intravascular

and the interstitial compartments. Only when concentrations proximal to the

Part I - General Introduction __________________________________________________

epithelium have risen sufficiently (relative to the Km value for its uptake) would active

accumulation occur at a significant rate. Since in the isolated perfused lung, the delivery

of PQ occurs through the vasculature, the sequence or pattern of exposure of lung cells

to PQ in this system may more closely resemble that occurring in vivo comparatively to

the lung slice model, in which the epithelium becomes directly exposed. The

observation that in vivo (Smith and Heath, 1974) the rate of accumulation (Vmax) of PQ

in the lung was only one-seventh of that found in vitro in lung slices led to a search for

compounds present in plasma and capable of blocking the uptake of PQ in the lung

(Lock et al., 1976). Subsequent to the identification of this transport system, a number

of naturally occurring amines have been identified, which competitively inhibit the

uptake of PQ into lung tissue and which themselves act as substrates and accumulated

in rat lung slices in a saturable manner, obeying Michaelis-Menten kinetics. These

amines include the diamines putrescine and cadaverine, the oligoamines spermidine and

spermine (Smith, 1982; Wyatt et al., 1988), and the disulfide cystamine (Lewis et al.,

1989) (Fig. 8). An important property of these specific polyamines is that they are

positively charged at a physiological pH, and, consequently, they have a high affinity

toward negatively charged cellular molecules. Thus polyamines are very soluble in

water, and they exert strong cation-anion interactions with macromolecules, mainly with

DNA and RNA (Marczynski, 1985), a feature that represents their best-known direct

physiological role in cellular functions such as cell growth, division, and differentiation

(Janne et al., 1978; Heby, 1981).

__________________________________________________Part I - General Introduction

N

H 2

C

H 3

N

H 2

N

H

+

N

H 2

N

H 2

0.702 nm

0.622 nm

N

H

N

+

N CH3 Fig. 8 - Chemical structure of paraquat (A) and putrescine (B), showing geometric

standards of the distance between N atoms (optimal distance to fit polyamine uptake

system is unknown). The chemical structure of cadaverin (C), spermidine (D) and

spermine (E) is also presented

A possible gene coding for a polyamine transporter (TPO1) was isolated from

eukaryotic cells and introduced into yeast cells (McNemar et al., 2001). Yeast cells

heterologously expressing TPO1, become sensitive to polyamines. For mammals,

though, it is not known whether the transporter(s) is (are) located at the apical or basal

side of the cells, and how the expression of the gene is regulated. The suggestion has

been put forward that the polyamines, which, as above mentioned, are known to

regulate cell growth, may play a role in the differentiation of alveolar epithelial type II

cells to type I cells (Smith, 1982). It has also been proposed that cystamine represents a

source of taurine, which may have an antioxidant role in the lung (Lewis et al., 1989).

Subjects in whom acute fulminant poisoning occurs generally ingested more than

40 mg of PQ ion/Kg of body weight (BW) (Vale et al., 1987). In these cases, the role of

the pulmonary transport system clearly has a negligible role on the evolution of the

NH 2

A

B

C

D

E

Part I - General Introduction __________________________________________________

intoxication, which progresses to multiorgan system failure. However, in cases of

moderate poisoning, where PQ plasma concentrations appear to be at the order of 10 to

20 μM, kinetic considerations suggest that only cells actively accumulating PQ would

achieve the intracellular concentrations necessary to cause significant cellular damage.

3.2.3 Structural requirements for the pulmonary polyamine uptake system

An important aim of earlier studies concerning pulmonary polyamine uptake

system (PUS) was to discover the structural requirements for substrates of the transport

system in order to find possible antagonists capable of preventing PQ from entering its

target cells. Ross and Krieger (Ross and Krieger, 1981) established that to act as a

substrate for the pulmonary PUS, a molecule must possess the following characteristics:

i) two or more positively charged nitrogen atoms, ii) maximum positivity of charge

surrounding these nitrogens, iii) a nonpolar group between these charges, and iv) a

minimum of steric hindrance. Gordonsmith et al. (Gordonsmith et al., 1983) have

demonstrated that the optimum distance (essential for binding and consequently, for

transport) between the nitrogen centers is four methylene groups (about 0.622 nm as it

occurs in putrescine), although a spacing between four and seven methylene groups is

tolerated. These assumptions explain how polyamines and PQ (with ≈ 0.702 between

two positively charged nitrogens) can share a common uptake system, but also why PQ

(with its steric hindrance of the nitrogens by the pyridine rings) is a less successful

substrate (Smith, 1987). The affinity of the uptake system for the polyamines appeared

to be sevenfold higher (i.e., exhibiting a lower apparent Km) than that of PQ (Smith,

1982). Although PQ proved to be a rather "poor" substrate (higher Km than polyamines)

for the PUS, it is undoubtedly "recognized" as a substrate, probably as a consequence of

its structural similarity to these endogenous substrates (Fig. 8), and is therefore

mistakenly accumulated into the lung, especially in the alveolar type I and II cells and in

the Clara cells, through this transport pathway (Smith, 1982). Later, O'Sullivan et al.

(O'Sullivan et al., 1991) showed that many putrescine analogues competitively inhibit

putrescine and PQ uptake. The authors established that the inhibition of putrescine

uptake by analogs decreases with increasing N-alkylation and those analogues with a

bulky substituent of the butyl chain do not inhibit the uptake at all. The strongest

inhibition was found with N-(4-aminobutyl)aziridine; this cytotoxic compound does not

__________________________________________________Part I - General Introduction

seem to alter the polyamine Vmax but might fit into the substrate binding site of the

receptor. The selective accumulation/retention of PQ in lung tissue provides a plausible

explanation for this organ selectivity to damage in comparison with other tissues.

Although the disposition of PQ in human tissues has not been as extensively studied as

in experimental animals, the major organs affected in man are also the lung and kidney.

Therefore, it seems likely that PQ is selectively accumulated in the human lung and

excreted by the kidney.

DQ exposure produce signs and symptoms similar to those of PQ except for one

important system - the pulmonary system (Jones and Vale, 2000). In contrast to PQ, DQ

is not a substrate for the pulmonary PUS and therefore is not selectively pneumotoxic.

In fact DQ exhibits a much smaller intramolecular distance between the two charged

nitrogen atoms, explaining its much greater safety margin (Rose and Smith, 1977).

3.2.4 Characterization of the pulmonary polyamine uptake system

It is clear that there will be a range of endogenous and exogenous compounds that

are capable of using this uptake system. Smith and Wyatt (Smith and Wyatt, 1981) and

Lewis et al. (Lewis et al., 1989) showed that the uptake of putrescine and cystamine in

rat lung slices was not dependent on the sodium (Na + ) concentration in the medium. In

contrast to these observations, Rannels et al. (Rannels et al., 1989) found that, in type II

pneumocytes the uptake of putrescine and spermidine was dependent on Na + , whereas

spermine uptake was not, indicating that polyamine uptake may take place via different

uptake systems. However, in these experiments, the nature and concentration of the ions

used to replace Na + were probably critical factors, because it has been shown that a

supplement of NaCl, LiCl, or choline significantly reduced the uptake of polyamines

due to the increase of osmotic pressure (Rannels et al., 1989). On the other hand,

Kumagai and Johnson (Kumagai and Johnson, 1988) showed that replacement of Na + by

mannitol or sucrose did not modulate putrescine uptake in rat enterocytes, whereas

replacement by choline, lithium (Li + ), N-methyl-D-glucamine, or tetramethylammonium

did. It was hypothesized that cations can interact with the carrier but that no co-transport

of Na + is involved in putrescine uptake. Another issue is whether there is one or more

pulmonary PUSs. In bovine arterial smooth muscle cells, Aziz et al. (Aziz et al., 1994)

and Jänne et al. (Janne et al., 1978) found that putrescine was accumulated through an

Part I - General Introduction __________________________________________________

uptake system that is also used by spermidine, spermine, PQ, and methylglyoxal bis-

(guanylhydrazone) (MGBG), but spermidine and spermine were also accumulated

through a different uptake system insensitive to putrescine and PQ, and only partially

sensitive to the presence of MGBG. Similarly, one study using suspensions of freshly

isolated type II pneumocytes (Chen et al., 1992) showed that putrescine uptake was

inhibited by PQ (and vice versa) in a partially competitive manner. These authors

postulated that the PUS in type II cells for PQ and putrescine possessed two separate

sites, one for each substrate, and that binding at one site leads to a conformational

change in the other. However, such partially competitive inhibition was not found in

other studies using hamster (Hoet et al., 1995) or human (Hoet et al., 1994) type II

pneumocytes. In another study performed in rat lung slices the inhibition of PQ

accumulation in presence of putrescine resulted from a process that appears to be

competitive (Karl and Friedman, 1983).

3.2.5 Cellular localization of the polyamine uptake system in the lung

The problem of the localization of the PUS in the lung was addressed first by

identifying the cellular targets for the toxicity of PQ and later by identifying the site of

accumulation of radiolabeled PQ and/or polyamines. Smith and Wyatt (Smith and

Wyatt, 1981) performed morphological and functional studies to localize the site of

cytotoxicity of PQ in lung slices taken from PQ (20 mg/Kg)-exposed rats. Lung slices

taken from rats 24 hours after treatment evidenced morphological damage to type I and

type II cells and their ability to take up putrescine (10 µM) or PQ (10 µM) was

impaired, thus suggesting that type I or type II pneumocytes are the site of uptake of

putrescine and PQ. Another experimental approach to determine the site of polyamine

uptake resulted from autoradiography. Waddell and Marlowe (Waddell and Marlowe,

1980) showed that after the i.v. administration of [ 14 C]PQ (10 µM) to mice, distribution

of the label corresponded to that in alveolar type II cells. Studies with rat lung slices by

Nemery et al. (Nemery et al., 1987) clearly demonstrated the presence of [ 3 H]putrescine

in alveolar type II cells and also in bronchiolar Clara cells (Fig. 9).

__________________________________________________Part I - General Introduction

Fig. 9 - Autoradiographs of rat lung tissue incubated with [ 3 H]putrescine. Resin sections

1 µm thick were stained with toluidine blue and examined by light microscopy.

Labeling occurs in alveolar walls and in alveolar type II pneumocytes (A and B, arrows).

There is no labeling in macrophages (B) or in walls of vessels, but Clara cells

(arrowheads) in bronchiolar epithelium (C) show intense labeling. Original

magnifications: ×600 in A; ×1,500 in B and C. Adapted from Nemery et al. (Nemery et

al., 1987).

Wyatt et al. (Wyatt et al., 1988), who carried out both in vivo and in vitro studies,

also showed uptake of [ 3 H]PQ, [ 3 H]putrescine, [ 3 H]spermidine, and [ 3 H]spermine by

alveolar type II cells and, at least in vitro, also by Clara cells. Hoet and co-workers also

visualized, by ultrastructural autoradiography, [ 14 C]putrescine in both type I and type II

cells of the alveolar epithelium, but not over the endothelium or any cells of the

Part I - General Introduction __________________________________________________

interstitium, in hamster (Hoet et al., 1995) and human (Hoet et al., 1993) lung slices

(Fig. 10).